In Situ Accessibility of Saccharomyces cerevisiae 26S rRNA to Cy3-Labeled Oligonucleotide Probes Comprising the D1 and D2 Domains

Abstract

In the last decade, fluorescence in situ hybridization (FISH) became the method of choice for the direct detection and identification of microorganisms in their natural environments (1, 3, 15). Even though FISH has been extensively used in ecological studies of bacteria (3) and other organisms (17), work with fungi has been restricted to the detection of Aureobasidium pullulans on the phylloplane (12, 19) and either clinically relevant or food spoilage yeasts (9, 10, 13, 14). Recently, a method using fluorescently labeled peptide nucleic acid probes was applied with success to the detection of Dekkera bruxellensis in wine (20), to the differentiation between Candida albicans and Candida dubliniensis (16), and to direct detection of C. albicans in blood culture bottles (18).

Preliminary studies with yeasts have shown that FISH assays are rapid and simple to carry out, do not require special cell permeation treatments and result in a high signal-to-noise ratio even when the cellular ribosome content is low, e.g., in late-stationary-phase cells (J. Inácio et al., unpublished data). However, a significant fraction of the probes designed yield low or no hybridization signals under optimal experimental conditions as assessed with a universal probe (10). One possible limitation of the method is associated with the target molecule, the rRNA. The targeted region of the ribosomes, which remain in the intact cell, might be structurally hindered or involved in molecular interactions, rendering it inaccessible to probe hybridization (3). Despite the development of procedures to improve the accessibility of those regions by using unlabeled helper oligonucleotides (6), a very useful clue when trying to design a good probe is to look for target sites located in rRNA regions already known to be accessible (7, 8).

The D1 and D2 domains at the 5′ end of 26S rRNA show a high degree of interspecies sequence variation for yeasts and are therefore frequently used for identification as well as in phylogenetic studies (5, 11). Due to the nucleotide sequence variability and to the large number of sequences available in public databases, this region provides an excellent basis to design species-specific FISH probes targeting the rRNAs of yeasts (16, 20).

The aim of the present study was to evaluate the accessibility of the D1 and D2 domains in the 26S rRNA to fluorescently labeled probes by using Saccharomyces cerevisiae as a model.

MATERIALS AND METHODS

Probe design.

Oligonucleotide probes were designed to cover the full length of the 26S rRNA D1 and D2 domains of S. cerevisiae (Fig. ​(Fig.1)1) (sequence retrieved from GenBank under accession number U44806). The sequences and positions of the 32 probes in the D1 and D2 domains are listed in Table ​Table1.1. The standard probe length of 18 nucleotides was varied if the estimated dissociation temperature (T_d), according to the formula of Suggs et al. (21) [T_d = 4 × (G + C) + 2 × (A + T)], exceeded 60°C or was below 48°C.

Open in a separate window

FIG. 1.

Fluorescence intensities of all oligonucleotide probes, standardized to that of the brightest probe (D-223), indicated in a model of the S. cerevisiae 26S rRNA secondary structure in which the D1 and D2 domains (delimited by the NL1 and NL4 primer target sites) are enlarged. The color coding indicates differences in the level of Cy3 probe-conferred fluorescence. The secondary structure is adapted from the European rRNA database (http://rrna.uia.ac.be).

TABLE 1.

Sequences, relative fluorescence intensities, and brightness classes of a set of Cy3-labeled oligonucleotide probes targeting the S. cerevisiae 26S rRNA D1 and D2 domains

Probe name	S. cerevisiae D1-D2 position (5′→3′)	Probe sequence (5′→3′)	Relative probe fluores- cence (%)a	Bright- ness class
D-1	1-20	AAGGCAATCCCGGTTGGTTT	67	II
D-21	21-39	CGCTTCACTCGCCGTTACT	71	II
D-40	40-58	TTCAAATTTGAGCTTTTGC	32	IV
D-59	59-77	GGCACCGAAGGTACCAGAT	15	V
D-78	78-96	CTCTCCAAATTACAACTCG	96	I
D-97	97-114	AACGGCCCCAAAGTTGCC	39	IV
D-115	115-132	CAAGGAACATAGACAAGG	4	VI
D-133	133-150	CTCTATGACGTCCTGTTC	81	I
D-151	151-168	CCACACGGGATTCTCACC	44	III
D-169	169-186	AAAGAACCGCACTCCTCG	66	II
D-187	187-204	TCTTCGAAGGCACTTTAC	17	V
D-205	205-222	ATTCCCAAACAACTCGAC	84	I
D-223	223-240	CCACCCACTTAGAGCTGC	100	I
D-241	241-259	TAGCTTTAGATGGAATTTA	18	V
D-260	260-278	TCGGTCTCTCGCCAATATT	66	II
D-279	279-297	TCACTGTACTTGTTCGCTA	79	II
D-298	298-316	AGTTCTTTTCATCTTTCCA	84	I
D-317	317-335	TTTTTCACTCTCTTTTCAA	74	II
D-336	336-354	TTTCAACAATTTCACGTAC	55	III
D-355	355-373	CTGATCAAATGCCCTTCCC	69	II
D-374	374-392	AGGGCACAAAACACCATGT	48	III
D-384	384-401	AAGGAGCAGAGGGCACAA	30	IV
D-402	402-419	CGAGATTCCCCTACCCAC	61	II
D-411	411-428	AGTGAAATGCGAGATTCC	13	V
D-429	429-446	CAAAACTGATGCTGGCCC	24	IV
D-447	447-464	ATGGATTTATCCTGCCAC	7	V
D-465	465-482	GAGGCAAGCTACATTCCT	16	V
D-483	483-500	CAGGCTATAATACTTACC	6	V
D-501	501-518	CAGCTGGCAGTATTCCCA	7	V
D-519	519-536	CGTCGCAGTCCTCAGTCC	83	I
D-537	537-554	GCCAGCATCCTTGACTTA	48	III
D-555	555-572	GCGGCATATAACCATTAT	55	III

Open in a separate window

^aFluorescence intensities expressed as a percentage of the value obtained for the brightest probe detected, D-223.

Comparison of 26S rRNA accessibilities in different yeasts.

To evaluate whether the accessibility data obtained for the region analyzed in the S. cerevisiae 26S rRNA could be extrapolated to other yeast species, a subset of the probes tested in this study were used in FISH experiments with several yeast species that presented a full complementary target site for those probes. The probes and yeast species selected are shown in Fig. ​Fig.2.2. The EUK 516 (5′-ACCAGACTTGCCCTCC) (2) and NonEUB (5′-ACTCCTACGGGAGGCAGC) (25) probes were used as positive and negative controls, respectively. All of the yeast strains were grown, harvested, and paraformaldehyde fixed as described above. The FISH experiments were carried out as indicated above, and 10 μl of the final hybridization mixture was spotted onto microscopic slides, air dried in the dark, and mounted with Vectashield solution (Vector, Burlingame, Calif.). The slides were examined with an Olympus BX50 microscope fitted for epifluorescence microscopy with a U-ULH 100-W mercury high-pressure bulb and a U-MA1007 filter set for the fluorochrome Cy3 (Olympus). The fluorescence intensity of the hybridization signal was checked visually. Photomicrographs were obtained with a digital camera (Olympus C3030-ZOOM) and edited with standard software (Adobe Photoshop 6.0).

Open in a separate window

FIG. 2.

Comparison of in situ hybridization signals for S. cerevisiae and other yeast species. The probes selected fall into different accessibility classes in the S. cerevisiae 26S rRNA D1 and D2 domains and have an identical target site in that region for all of the yeasts indicated.

RESULTS AND DISCUSSION

The results obtained for the in situ accessibility of S. cerevisiae 26S rRNA to Cy3-labeled oligonucleotide probes covering the full length of the D1 and D2 domains are shown in Fig. ​Fig.11 and Table ​Table1.1. Fluorescence intensities for each probe were quantified by flow cytometry, expressed as a percentage of the fluorescence signal of the brightest probe detected (D-223), and grouped into different accessibility classes (7). The fluorescence intensity obtained for probe D-223 was of the same order of magnitude as the signal shown by the universal eukaryote probe EUK 516, which is targeted to the 18S rRNA. About 44% of the probes tested belong to the higher-accessibility classes (I and II), and 28% were poorly binding (brightness classes V and VI). To evaluate whether the probes belonging to the most inaccessible classes (IV, V, and VI) would show better fluorescent signals under different hybridization conditions, a subset of these probes was chosen and hybridization reactions were performed at different temperatures. The use of hybridization conditions with different stringencies did not significantly improve the fluorescence intensities (data not shown), in accordance with previous studies (7). The overall results indicate that, despite its short length of approximately 600 nucleotides, the D1 and D2 domains include potentially good targets for yeast probe design. However, care should be taken when selecting target sites complementary to the most variable areas of the D1 and D2 domains (Fig. ​(Fig.3),3), where it is easier to find species-specific sequences. The data obtained show that the most conserved stretches of the studied region are more accessible (Fig. ​(Fig.3)3) (e.g., positions 200 to 350), and the most variable areas often show medium to low accessibility (e.g., the region between nucleotides 415 and 510). A similar trend has been observed in a previous accessibility study conducted for Escherichia coli 16S rRNA (7).

Open in a separate window

FIG. 3.

Comparison of the relative in situ accessibilities (black line) of the S. cerevisiae 26S rRNA D1 and D2 domains and the average nucleotide substitution rates (gray) in yeasts.

As for other probes belonging to the weaker accessibility classes (IV, V, and VI), whose low probe-conferred fluorescence signals may be due to the rRNA secondary structure and/or protein-rRNA interactions, the weaker signal of probe D-59 can additionally be attributed to a significant degree of self-complementarity in its sequence, an aspect to be taken into consideration when designing species-specific probes. Another possible explanation for the less intense fluorescence signals observed with some FISH probes is the quenching due to the presence of guanine residues adjacent to the 3′ ends of the rRNA targets (4, 22). We observed no significant correlation between the probe-conferred fluorescence intensities and this nucleobase position in the vicinity of the 3′ ends of the respective rRNA target sites (data not shown). This observation agrees with those of Torimura and colleagues (22), who observed the quenching phenomenon for fluorescein isothiocyanate-labeled oligonucleotides but not for Cy3-labeled ones.

Interestingly, a comparative analysis of the in situ accessibility of the first 350 nucleotides in E. coli 23S rRNA to Cy3-labeled oligonucleotide probes (8) and the data obtained in this work for S. cerevisiae show some striking similarities (Fig. ​(Fig.4).4). Although the probes used have different target sequences in the two microorganisms, the accessibilities follow the same general trend. On the other hand, the probes belonging to the higher accessibility classes (I and II) in S. cerevisiae also yielded strong hybridization signals with species belonging to different phylogenetic groups, including the distantly related basidiomycetous yeasts (Fig. ​(Fig.2).2). This suggests that the D1-D2 accessibility map presented here for S. cerevisiae provides useful guidance for the design of species-specific probes for other yeasts, maybe even for other fungi or eukaryotic microorganisms. However, the design of probes for more distantly related organisms would probably require a different model.

Open in a separate window

FIG. 4.

Comparison of the accessibilities of homologous regions in S. cerevisiae 26S rRNA and E. coli 23S rRNA to Cy3-labeled probes.

With this study we hope to contribute to a more rational design of fluorescently labeled probes for yeast identification that will stimulate the use of FISH-based methods in a wide range of applications, including studies on the ecology of yeasts.

Acknowledgments

REFERENCES